Improved computing performance and load balancing of atmospheric general circulation model

Abstract

The current situation is unprecedented in the history of the Earth, as emissions of greenhouse gases could increase the mean global air temperature over several decades, while the natural temperature increasing by the same amount will take no less than several millennia. Therefore, to carry out basic research  on this problem we must study the appropriate mathematical models. The climatemodel of the Computing Center, Russian Academy of Sciences also proved quite effective as general circulation models of the atmosphere, ocean model, etc. However, these models still contains some complete issues. The purpose of this paper is to modify the parallel algorithm used in general circulation models of the atmosphere (AGCM); thereby improve computing performance. Especially the exploitation of load balancing in case there are many compute nodes and the computing resources are heterogeneous. The improvement shown by the concurrent exploitation of two processor groups are corresponding to two blocks of physics and dynamics on the same data. The results are also tested on the experimental  data, and hence show the effectiveness of the improved algorithm.